Impact Quantification Research Articles

Deep learning based carbon emissions forecasting and renewable energy's impact quantification

Global warming is one of the most challenging issues of the current era. Revolutions in industrial, Information and Communication Technology (ICT) sectors significantly contribute in increasing global warming. Green House Gases (GHG) emissions from industrial, transportation, power and other sectors cause environmental pollution, which results in climate degradation. Environmental experts are well aware of the disastrous consequences of excessive global warming; therefore, several decarbonization strategies are developed and practiced in the recent past. A widely practiced decarbonization strategy is replacing fossil fuels by Renewable Energy Sources (RES) in the power systems. Electricity consumers are also encouraged to shift their consumption load to low carbon emissions' time periods. For accomplishing this task, an estimation of future carbon emissions is required. In this paper, power system's carbon emissions are predicted accurately with the help of a novel and an efficient forecasting model. The proposed model comprises of Spearman Correlation Analysis (SCA) and Improved Shallow Denoising Autoencoder (ISDAE) based feature engineering. Forecasting is performed through an Improved Particle Swarm Optimization (IPSO) based Deep Neural Network (DNN) forecaster. In addition, a comprehensive quantification analysis is also presented in this paper. The impacts of RES integration level on the electricity price, consumption cost and Greenhouse Gases (GHG) emissions are quantified descriptively and graphically. Performance of the proposed forecasting model is evaluated by Normalized Root Mean Square Error (NRMSE), Mean Absolute Error (MAE) and Mean Square Error (MSE). Simulation results prove that the proposed model outperforms Support Vector Machine (SVM) and Multiple Linear Regression (MLR) based carbon emission forecasting models in terms of forecasting accuracy.

Evaluating Metrics for Impact Quantification

Evaluating Metrics for Impact Quantification

OpenConcrete: a tool for estimating the environmental impacts from concrete production

As the increasing global consumption of concrete drives notable environmental burdens from its production, particularly greenhouse gas (GHG) emissions, interest in mitigation efforts is increasing. Yet current environmental impact quantification tools rely on user decision-making to select data for each concrete constituent, have inconsistent scopes and system boundaries, and often utilize third-party life cycle inventories. These factors limit customization or tracking of data and hinder the ability to draw robust comparisons among concrete mixtures to mitigate its environmental burdens. To address these issues, we introduce a cohesive, unified dataset of material, energy, and emission inventories to quantify the environmental impacts of concrete. In this work, we detail the synthesis of this open dataset and create an environmental impact assessment tool using this data. Models can be customized to be region specific, expanded to varying concrete mixtures, and support data visualization throughout each production stage. We perform a scenario analysis of impacts to produce a representative concrete mixture across the United States, with results ranging from 189 kg CO2-eq/m3 of concrete (California) to 266 kg CO2-eq/m3 of concrete (West Virginia). The largest driver of GHG, nitrogen oxide, sulfur oxide, and volatile organic compound emissions as well as energy demand is cement production, but aggregate production is the largest driver of water consumption and particulate matter smaller than 2.5 microns (PM2.5) emissions.

Graphical Ways to Visualize Operational Risk Results for Transmission System Contingencies

The increased complexity of the transmission grid can endanger the operational security of the grid. Operational risk assessment, a stochastic tool, helps to enhance security. Contingency analysis and its impact quantification are the main constituents of operational risk assessment. In this study, different graphical methods are proposed to visualize operational risk contingency-based detailed results: heat-map and risk-based contingency chart. Through the heat-map, the system operator can determine which contingencies contribute most to the operational risk and would therefore be the most threatening contingencies for operational security of the grid. The “risk-based contingency chart” allows the system operator to analyze contingency cases from the probability and impact aspect in one chart. Both tools may be used in the control room for improved operational planning. In this study of contingency analysis and various types of network studies of severity factor quantification, the IEEE 39-Bus sample network is used in Power-Factory to analyze the contingencies behavior under different operational scenarios.

A new curb lane monitoring and illegal parking impact estimation approach based on queueing theory and computer vision for cameras with low resolution and low frame rate

The rapid development of the internet of things (IoT), sensing technologies, and machine learning and deep learning techniques, along with the growing variety and volume of data, have yielded new perspectives on how novel technologies can be applied to obtain new sources of curb data to achieve cost-effective curb management. This study presents a new computer vision–based data acquisition and analytics approach for curb lane monitoring and illegal parking impact assessment. The proposed “rank, detect, and quantify impacts” system consists of three main modules: 1) hotspot identification based on rankings generated by local spatial autocorrelation analysis, 2) curb lane occupancy estimation leveraging traffic cameras and computer vision techniques, and 3) illegal parking traffic impact quantification using an M/M/∞ queueing model. To demonstrate the feasibility and validity of the proposed approach, it was tested and empirically validated using field data collected from three case study sites in Midtown Manhattan, New York City (NYC)—one of the most complex urban transportation networks in the world. Different types of curb lane occupancy, including parking and bus lanes, and different frequencies of illegal parking (high, moderate, low) were investigated. The proposed method was proven to be effective even for low resolution and discontinuous video feeds obtained from publicly available traffic cameras. All three case study sites achieved good detection accuracy (86% to 96%) for parking and bus lane occupancy, and acceptable precision and recall in detecting illegal parking events. The queueing model was also proven to effectively quantify link travel time with the appearance of illegal parking events of different frequencies. The proposed “rank, detect, and quantify impacts” system is friendly for large-scale real-time implementation and is highly scalable to help evaluate the impact of other modes such as bike or mobility-on-demand (MOD) services. It can also be easily adopted by other cities to provide transportation agencies with effective data collection and innovative curb space management strategies.

Weather impact quantification on airport arrival on-time performance through a Bayesian statistics modeling approach

Compared with departures, predicting the weather impact on arrival delays is more challenging because of possible non-linear, cascading effects, and higher uncertainty. Existing weather impact studies are location-dependent and often neglect the impacts of dangerous phenomena. We propose a data-driven model for severe weather impact quantification on airport arrival on-time performance based on the Bayesian approach to address these issues. Our model considers the impact of the dangerous phenomenon by evaluating the mean shift and is flexible enough to be applied to different airports. Using two years’ worth of data (2017-2018) from the Hong Kong International Airport, we studied over 55,000 local meteorological reports and analyzed over 430,000 arrival flights. Across all three key performance metrics considered, a non-linear relationship with the weather score, akin to a phase transition, could be observed. This framework allows a comparison between the sensitivity of each airport’s arrival performance metric towards severe weather. Delay rate is the most sensitive metric, while cancellation rate is the least. For the impacts of dangerous phenomena, cumulonimbus has the most significant impact on the delay rate. Shower rainfall/cumulonimbus has a similar and vital impact on the mean arrival delay per hour. Because of its potential applications in different airports, this framework can provide a deeper insight into weather impact on air traffic networks.

ScEpiLock: A Weakly Supervised Learning Framework for cis-Regulatory Element Localization and Variant Impact Quantification for Single-Cell Epigenetic Data.

Recent advances in single-cell transposase-accessible chromatin using a sequencing assay (scATAC-seq) allow cellular heterogeneity dissection and regulatory landscape reconstruction with an unprecedented resolution. However, compared to bulk-sequencing, its ultra-high missingness remarkably reduces usable reads in each cell type, resulting in broader, fuzzier peak boundary definitions and limiting our ability to pinpoint functional regions and interpret variant impacts precisely. We propose a weakly supervised learning method, scEpiLock, to directly identify core functional regions from coarse peak labels and quantify variant impacts in a cell-type-specific manner. First, scEpiLock uses a multi-label classifier to predict chromatin accessibility via a deep convolutional neural network. Then, its weakly supervised object detection module further refines the peak boundary definition using gradient-weighted class activation mapping (Grad-CAM). Finally, scEpiLock provides cell-type-specific variant impacts within a given peak region. We applied scEpiLock to various scATAC-seq datasets and found that it achieves an area under receiver operating characteristic curve (AUC) of ~0.9 and an area under precision recall (AUPR) above 0.7. Besides, scEpiLock’s object detection condenses coarse peaks to only ⅓ of their original size while still reporting higher conservation scores. In addition, we applied scEpiLock on brain scATAC-seq data and reported several genome-wide association studies (GWAS) variants disrupting regulatory elements around known risk genes for Alzheimer’s disease, demonstrating its potential to provide cell-type-specific biological insights in disease studies.

The Microsoft Academic Knowledge Graph enhanced: Author name disambiguation, publication classification, and embeddings

Abstract Although several large knowledge graphs have been proposed in the scholarly field, such graphs are limited with respect to several data quality dimensions such as accuracy and coverage. In this article, we present methods for enhancing the Microsoft Academic Knowledge Graph (MAKG), a recently published large-scale knowledge graph containing metadata about scientific publications and associated authors, venues, and affiliations. Based on a qualitative analysis of the MAKG, we address three aspects. First, we adopt and evaluate unsupervised approaches for large-scale author name disambiguation. Second, we develop and evaluate methods for tagging publications by their discipline and by keywords, facilitating enhanced search and recommendation of publications and associated entities. Third, we compute and evaluate embeddings for all 239 million publications, 243 million authors, 49,000 journals, and 16,000 conference entities in the MAKG based on several state-of-the-art embedding techniques. Finally, we provide statistics for the updated MAKG. Our final MAKG is publicly available at https://makg.org and can be used for the search or recommendation of scholarly entities, as well as enhanced scientific impact quantification.

A fully-printed electrochemical platform for assisted colorimetric detection of phosphate in saliva: Greenness and whiteness quantification by the AGREE and RGB tools

Herein, we report the environmental impact quantification of a newly developed fully printed electrochemical device to assist a colorimetric detection of phosphate in saliva. The evaluation of the analytical procedure was performed according to the principles of Green Analytical Chemistry and White Analytical Chemistry. The standard method for phosphate detection relies on a reaction between phosphate and molybdate in presence of antimony potassium tartrate and ascorbic acid, using strong acid conditions and high volumes of reagents (100–500 mL). To deliver an eco-friendly method, we have combined a screen-printed electrode with a liquid electrolyte battery and inkjet-printed conductive paths to develop a fully printed device on a flexible polymer substrate avoiding the use of ascorbic acid and using a small amount of reagents. The printed sensor was first developed and optimized for phosphate detection in saliva, allowing for a detection limit equal to 26 µM and satisfactory repeatability (relative standard deviation value of 7.5%). Finally, the AGREE and the RGB assessment tools were applied for a quantitative evaluation of the proposed sensor and reference method, in agreement with the Green Analytical and White Analytical principles. The results demonstrated the lower environmental impact of the proposed sensor, as well as the suitability of this novel approach for phosphate detection in saliva.

InsuLock: A Weakly Supervised Learning Approach for Accurate Insulator Prediction, and Variant Impact Quantification.

Mapping chromatin insulator loops is crucial to investigating genome evolution, elucidating critical biological functions, and ultimately quantifying variant impact in diseases. However, chromatin conformation profiling assays are usually expensive, time-consuming, and may report fuzzy insulator annotations with low resolution. Therefore, we propose a weakly supervised deep learning method, InsuLock, to address these challenges. Specifically, InsuLock first utilizes a Siamese neural network to predict the existence of insulators within a given region (up to 2000 bp). Then, it uses an object detection module for precise insulator boundary localization via gradient-weighted class activation mapping (~40 bp resolution). Finally, it quantifies variant impacts by comparing the insulator score differences between the wild-type and mutant alleles. We applied InsuLock on various bulk and single-cell datasets for performance testing and benchmarking. We showed that it outperformed existing methods with an AUROC of ~0.96 and condensed insulator annotations to ~2.5% of their original size while still demonstrating higher conservation scores and better motif enrichments. Finally, we utilized InsuLock to make cell-type-specific variant impacts from brain scATAC-seq data and identified a schizophrenia GWAS variant disrupting an insulator loop proximal to a known risk gene, indicating a possible new mechanism of action for the disease.

Data availability and sector‐specific frameworks restrict drought impact quantification in the Intermountain West

AbstractAs is the case for many semi‐arid regions globally, drought in the Intermountain West of the United States is a recurrent, costly phenomenon that leaves few aspects of human and natural systems untouched. Here, we focus on drought impact data and evaluation challenges across four non‐agricultural sectors: water utilities, forest resources, public health, and recreation and tourism. There are marked commonalities in the way drought indicators—that is, hydrometeorological conditions—are tracked, but considerable differences in how impacts are measured, evaluated, and disseminated. For drought indicator data, researchers and practitioners have a veritable smorgasbord of data at their fingertips. Such data are often spatially and temporally continuous, available at a wide variety of scales, and readily accessible through government‐funded online portals. This is in stark contrast to drought impact data, which are typically collected opportunistically, if at all. These data are thus often limited in spatiotemporal scope and difficult to access relative to drought indicators. Concerningly, even within a given sector, the definition of drought impacts, quantitative or otherwise, can vary considerably, making it difficult to evaluate the true cost of drought. Far from being specific to the Intermountain West, these problems are found in most regions experiencing drought. We suggest such challenges are surmountable through the development of a common drought impact framework based around economic damages and purposeful, continuous, government‐funded drought impact data collection. These tractable changes will allow for a better quantification of drought's true impacts under both present conditions and climate change scenarios in the Intermountain West and beyond.This article is categorized under: Human Water > Value of Water Science of Water > Water Extremes Water and Life > Stresses and Pressures on Ecosystems

Sovereign Digital Consent through Privacy Impact Quantification and Dynamic Consent

Digitization is becoming more and more important in the medical sector. Through electronic health records and the growing amount of digital data of patients available, big data research finds an increasing amount of use cases. The rising amount of data and the imposing privacy risks can be overwhelming for patients, so they can have the feeling of being out of control of their data. Several previous studies on digital consent have tried to solve this problem and empower the patient. However, there are no complete solution for the arising questions yet. This paper presents the concept of Sovereign Digital Consent by the combination of a consent privacy impact quantification and a technology for proactive sovereign consent. The privacy impact quantification supports the patient to comprehend the potential risk when sharing the data and considers the personal preferences regarding acceptance for a research project. The proactive dynamic consent implementation provides an implementation for fine granular digital consent, using medical data categorization terminology. This gives patients the ability to control their consent decisions dynamically and is research friendly through the automatic enforcement of the patients’ consent decision. Both technologies are evaluated and implemented in a prototypical application. With the combination of those technologies, a promising step towards patient empowerment through Sovereign Digital Consent can be made.

Model-Driven Impact Quantification of Energy Resource Redundancy and Server Rejuvenation on the Dependability of Medical Sensor Networks in Smart Hospitals.

The spread of the Coronavirus (COVID-19) pandemic across countries all over the world urges governments to revolutionize the traditional medical hospitals/centers to provide sustainable and trustworthy medical services to patients under the pressure of the huge overload on the computing systems of wireless sensor networks (WSNs) for medical monitoring as well as treatment services of medical professionals. Uncertain malfunctions in any part of the medical computing infrastructure, from its power system in a remote area to the local computing systems at a smart hospital, can cause critical failures in medical monitoring services, which could lead to a fatal loss of human life in the worst case. Therefore, early design in the medical computing infrastructure’s power and computing systems needs to carefully consider the dependability characteristics, including the reliability and availability of the WSNs in smart hospitals under an uncertain outage of any part of the energy resources or failures of computing servers, especially due to software aging. In that regard, we propose reliability and availability models adopting stochastic Petri net (SPN) to quantify the impact of energy resources and server rejuvenation on the dependability of medical sensor networks. Three different availability models (A, B, and C) are developed in accordance with various operational configurations of a smart hospital’s computing infrastructure to assimilate the impact of energy resource redundancy and server rejuvenation techniques for high availability. Moreover, a comprehensive sensitivity analysis is performed to investigate the components that impose the greatest impact on the system availability. The analysis results indicate different impacts of the considered configurations on the WSN’s operational availability in smart hospitals, particularly 99.40%, 99.53%, and 99.64% for the configurations A, B, and C, respectively. This result highlights the difference of 21 h of downtime per year when comparing the worst with the best case. This study can help leverage the early design of smart hospitals considering its wireless medical sensor networks’ dependability in quality of service to cope with overloading medical services in world-wide virus pandemics.

A new protocol for estimation of woody aboveground biomass in disturbance-prone ecosystems

Almost one third of global drylands are open forests and savannas, which are typically shaped by frequent natural disturbances such as wildfire and herbivory. Studies on ecosystem functions and services of woody vegetation require robust estimates of aboveground biomass (AGB). However, most methods have been developed for comparatively undisturbed forest ecosystems. As they are not tailored to accurately quantify AGB of small and irregular growth forms, their application on these growth forms may lead to unreliable or even biased AGB estimates in disturbance-prone dryland ecosystems. Moreover, these methods cannot quantify AGB losses caused by disturbance agents. Here we propose a methodology to estimate individual- and stand-level woody AGB in disturbance-prone ecosystems. It consists of flexible field sampling routines and estimation workflows for six growth classes, delineated by size and damage criteria. It also comprises a detailed damage assessment, harnessing the ecological archive of woody growth for past disturbances. Based on large inventories collected along steep gradients of elephant disturbances in African dryland ecosystems, we compared the AGB estimates generated with our proposed method against estimates from a less adapted forest inventory method. We evaluated the necessary stepwise procedures of method adaptation and analyzed each step’s effect on stand-level AGB estimation. We further explored additional advantages of our proposed method with regard to disturbance impact quantification. Results indicate that a majority of growth forms and individuals in savanna vegetation could only be assessed if methods of AGB estimation were adapted to the conditions of a disturbance-prone ecosystem. Furthermore, our damage assessment demonstrated that one third to half of all woody AGB was lost to disturbances. Consequently, less adapted methods may be insufficient and are likely to render inaccurate AGB estimations. Our proposed method has the potential to accurately quantify woody AGB in disturbance-prone ecosystems, as well as AGB losses. Our method is more time consuming than conventional allometric approaches, yet it can cover sufficient areas within reasonable timespans, and can also be easily adapted to alternative sampling schemes.

Effect of Temperature on Sowing Dates of Wheat under Arid and Semi-Arid Climatic Regions and Impact Quantification of Climate Change through Mechanistic Modeling with Evidence from Field

Rising temperature from climate change is the most threatening factor worldwide for crop production. Sustainable wheat production is a challenge due to climate change and variability, which is ultimately a serious threat to food security in Pakistan. A series of field experiments were conducted during seasons 2013–2014 and 2014–2015 in the semi-arid (Faisalabad) and arid (Layyah) regions of Punjab-Pakistan. Three spring wheat genotypes were evaluated under eleven sowing dates from 16 October to 16 March, with an interval of 14–16 days in the two regions. Data for the model calibration and evaluation were collected from field experiments following the standard procedures and protocols. The grain yield under future climate scenarios was simulated by using a well-calibrated CERES-wheat model included in DSSAT v4.7. Future (2051–2100) and baseline (1980–2015) climatic data were simulated using 29 global circulation models (GCMs) under representative concentration pathway (RCP) 8.5. These GCMs were distributed among five quadrants of climatic conditions (Hot/Wet, Hot/Dry, Cool/Dry, Cool/Wet, and Middle) by a stretched distribution approach based on temperature and rainfall change. A maximum of ten GCMs predicted the chances of Middle climatic conditions during the second half of the century (2051–2100). The average temperature during the wheat season in a semi-arid region and arid region would increase by 3.52 °C and 3.84 °C, respectively, under Middle climatic conditions using the RCP 8.5 scenario during the second half-century. The simulated grain yield was reduced by 23.5% in the semi-arid region and 35.45% in the arid region under Middle climatic conditions (scenario). Mean seasonal temperature (MST) of sowing dates ranged from 16 to 27.3 °C, while the mean temperature from the heading to maturity (MTHM) stage was varying between 12.9 to 30.4 °C. Coefficients of determination (R2) between wheat morphology parameters and temperature were highly significant, with a range of 0.84–0.96. Impacts of temperature on wheat sown on 15 March were found to be as severe as to exterminate the crop before heading. The spikes and spikelets were not formed under a mean seasonal temperature higher than 25.5 °C. In a nutshell, elevated temperature (3–4 °C) till the end-century can reduce grain yield by about 30% in semi-arid and arid regions of Pakistan. These findings are crucial for growers and especially for policymakers to decide on sustainable wheat production for food security in the region.

The role of data variability and uncertainty in the probability of mitigating environmental impacts from cement and concrete

Concrete is the most produced manmade material globally. This widespread production results in significant anthropogenic environmental impacts, the awareness of which has spurred advances in material development to lower these burdens. However, proposed changes are often not assessed in the context of the data variability and uncertainty inherent in the environmental impact quantification methods employed. As such, the probability that any suggested strategy will result in a desired effect is not addressed. This work aims to quantitatively examine data variability, an inherent characteristic of elements in supply chains, and data uncertainty, a function of data quality for the system being modeled, in assessments of greenhouse gas (GHG) and air pollutant emissions from concrete production. Data variability is determined through ranges in requisite input values from the literature; data uncertainty is assessed through application of an established pedigree matrix method. Statistical analysis of the emissions from concrete production incorporating sources of variability and uncertainty are examined through Monte Carlo simulations. Concrete mixtures, representing a feasible structural concrete for use in California infrastructure and three alternative mixtures are assessed, as are three GHG emissions mitigation strategies, namely, a change in thermal energy fuel mix, a change in electricity grid, and use of carbon capture and storage. The distributions of emissions derived through statistical analyses are used to examine the probability of efficacy of these strategies, as well as potential co-benefits on air emissions. Results show each constituent change and each mitigation strategy considered would lead to a reduction in GHG emissions if only mean values are compared; however, the probability of these reductions varies. These findings suggest mitigation efforts may not be as definitive as current assessments suggest. Results indicate the importance of using statistical methods to target desirable mitigation efforts in the environmental impacts from concrete production.

Climate change, marine resources and a small Chilean community: making the connections

Climate change is affecting large-scale oceanic processes. How and when these changes will impact those reliant on marine resources is not yet clear. Here we use end-to-end modeling to track the impacts of expected changes through the marine ecosystem on a specific, small community: Cochamó, in the Gulf of Ancud wider area, Chile. This area is important for Chilean fisheries and aquaculture, with Cochamó reliant on both lower and upper trophic level marine resources. We applied the GOTM-ERSEM-BFM coupled hydro-biogeochemical water-column model to gauge lower-trophic level marine ecological community response to bottom-up stressors (climate change, ocean acidification), coupled to an existing Ecopath with Ecosim model for the area, which included top-down stressors (fishing). Social scientists also used participatory modeling (Systems Thinking and Bayesian Belief Networking) to identify key resources for Cochamó residents and to assess the community’s vulnerability to possible changes in key resources. Modeling results suggest that flagellate phytoplankton abundance will increase at the cost of other species (particularly diatoms), resulting in a greater risk of harmful algae blooms. Both climate change and acidification slightly increased primary production in the model. Higher trophic level results indicate that some targeted pelagic resources will decline (while benthic ones may benefit), but that these effects might be mitigated by strong fisheries management efforts. Participatory modeling suggests that Cochamó inhabitants anticipate marine ecosystem changes but are divided about possible adaptation strategies. For climate change impact quantification, detailed experimental studies are recommended based on the dominant threats identified here, with specific local species.

BIM-DLCA: An integrated dynamic environmental impact assessment model for buildings

Life cycle assessment (LCA) is an important and widely used environmental impact quantification methodology. Recently, temporal variations have generated renewed interests in LCA, with dynamic LCA (DLCA) becoming a new research trend. Traditional data collection approaches are considerably time and effort intensive, with major challenges associated with the acquisition of high quality data that are available. This study integrates building information modeling (BIM) with DLCA to develop a new assessment model for buildings. It follows the basic LCA paradigm and includes five parts: goal and scope definition, BIM module, dynamic database, dynamic assessment, and interpretation. In the BIM-DLCA model, BIM was used to extract material information and the construction schedule. A number of analysis software and simulation tools were used to quantify the temporal elementary flows throughout the lifecycle of the building. Various forms of dynamic data were aggregated to create a database that supports dynamic assessment. A case study was used to testify the BIM-DLCA model and the case study's outcomes indicate that BIM has the potential to simplify data collection process and support DLCA. The proposed BIM-DLCA model is operable in practice. With the rapid development of digitalization and intelligence, the proposed BIM-DLCA model lays an important foundation for future assessment studies and promotes smart LCA.

Sustainability assessment of levulinic acid and succinic acid production from empty fruit bunch

Empty fruit bunch (EFB) generated as waste in plantation mill activities in Malaysia is a potential biomass feedstock of biorefinery for fuels and chemicals production. Levulinic acid and succinic acid, two out of 12 chemical building blocks identified by Department of Energy (DOE) to be used in synthesis of high-value materials, can be produced from biochemical conversion of the EFB. This paper evaluates sustainability assessment of EFB to levulinic acid and succinic acid. The assessments include net present value (NPV), global warming potential (GWP), and Hazard identification and ranking (HIRA) to cater for economic, environment, and safety performances, respectively. The results show that the levulinic acid production is more economically attractive than succinic acid production. The environmental impact quantification reveals that the levulinic acid has lower GWP score of 6.3 kgCO2-eq/kg levulinic acid than succinic acid with 11.2 kgCO2-eq/kg succinic acid. Meanwhile, the succinic acid production is inherently safer than levulinic acid production due to its less severe operating conditions.

Cooperative driver pathway discovery via fusion of multi-relational data of genes, miRNAs and pathways.

Discovering driver pathways is an essential step to uncover the molecular mechanism underlying cancer and to explore precise treatments for cancer patients. However, due to the difficulties of mapping genes to pathways and the limited knowledge about pathway interactions, most previous work focus on identifying individual pathways. In practice, two (or even more) pathways interplay and often cooperatively trigger cancer. In this study, we proposed a new approach called CDPathway to discover cooperative driver pathways. First, CDPathway introduces a driver impact quantification function to quantify the driver weight of each gene. CDPathway assumes that genes with larger weights contribute more to the occurrence of the target disease and identifies them as candidate driver genes. Next, it constructs a heterogeneous network composed of genes, miRNAs and pathways nodes based on the known intra(inter)-relations between them and assigns the quantified driver weights to gene-pathway and gene-miRNA relational edges. To transfer driver impacts of genes to pathway interaction pairs, CDPathway collaboratively factorizes the weighted adjacency matrices of the heterogeneous network to explore the latent relations between genes, miRNAs and pathways. After this, it reconstructs the pathway interaction network and identifies the pathway pairs with maximal interactive and driver weights as cooperative driver pathways. Experimental results on the breast, uterine corpus endometrial carcinoma and ovarian cancer data from The Cancer Genome Atlas show that CDPathway can effectively identify candidate driver genes [area under the receiver operating characteristic curve (AUROC) of $\geq $0.9] and reconstruct the pathway interaction network (AUROC of>0.9), and it uncovers much more known (potential) driver genes than other competitive methods. In addition, CDPathway identifies 150% more driver pathways and 60% more potential cooperative driver pathways than the competing methods. The code of CDPathway is available at http://mlda.swu.edu.cn/codes.php?name=CDPathway.